Language as Metaphor and Metonym in Ethnographic Literature

Michael Mahrt

Anthropology has from its start been concerned with the question of contextuality. From the beginning anthropology has claimed that the apparent irrational, contradictory or plainly wrong beliefs of primitive peoples were indeed understandable and explainable if they were seen in their own context and not in the context of the western beliefs. Anthropology is thus born of the enlightenment paradox between liberalism and the grand reason (Gellner ??). In modern terms this paradox has been stated as the more general problem of relativism versus universalism. As Gellner also observes in his important article, the emphasis on contextualism in anthropology has led to certain preconceptions. It has had the consequence, that anthropologists present their analysis as if there is no such thing as incoherence or contradictions in other cultures [1]. Anything can be explained, and indeed anything is explained with reference to contextuality.

The insistence on contextuality is indisputably one of anthropology’s great contributions. But it can be fallacious. In a sense it can be said, that what Gellner warns against is the tendency of absolutism in this preoccupation with contextuality. Although this problem has been pointed out by various authors for many years, it is nevertheless still central to the discipline. My attempt in this paper is to explore some of the underlying premises for the continuance of this problem of modern anthropology.

Although I am primarily concerned with academic anthropology in this paper, I do believe it is of relevance to other disciplines concerned with the study of other peoples, and especially to the problem of the power of representing otherness as the others .

One of the major reasons, in my view, that the problem of absolutism is still prevalent in modern anthropology, is that the discipline seldom questions the configuration of its own object. Whereas its nature is continually contested, the possibility of the existence of the ethnographic object is seldom explored. It is taken for granted that the object has a nature to be contested or questioned. There are many reasons for this. One of them is the insistence on fieldwork and the subsequent ethnographic monograph. Now, the problem is not fieldwork as such, since it is exactly through fieldwork that anthropology has one of its great potentials of contributing to the sciences. As Bruce Kapferer continues to point out, it is through meticulous fieldwork, that anthropology learns about what it (also) is like to be human. (Kapferer personal communication). I shall come back to this point later on. So the problem is not fieldwork as such, but rather, in my view, that fieldwork must result in ethnographic monographs. Monographs put their emphasis on the context within the object - that is a people - and maybe even the objects context of other peoples or the modern worldsystem, or what have you, but rarely the very justification of the object itself. It may be said, that this is exactly what post-modern critique has been about in anthropology, but, as I will show, it is in fact the very foundation of the post-modern critique itself.

What I propose to do is to question the nature of the context, that anthropology insists on, by exploring the configuration of the anthropological object. Since this configuration is ultimately a question of classification, I will start with a brief outline of Michel Foucault’s idea of what constitutes the possibility of a classification system.

Classification

Classification is a central subject in ethnographic literature. From the beginning of the systematic study of other peoples, the peculiar systems of kinship classification, botanical and zoological classification and so forth, has caught the interest of a great many ethnographers. But classification is also practised by the ethnographer in the ethnographic literature itself. Ethnography is based on a classification system of peoples. The part played by language in the logic of this classification system is my concern in this paper.

In ”The Order of Things” Michel Foucault analyses a classification of animals which is found in the writings of Jorge Luis Borges:

”This passage quotes a ‘certain Chinese encyclopaedia’ in which it is written that ‘animals are divided into (a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (l) et cetera , (m) having just broken the water pitcher, (n) that from a long way off look like flies’”. (Foucault [1966] 1992:xv).

What concerns Foucault is not the peculiarity of the individual categories, but the impossibility of the classification system itself. None of the individual categories are impossible in themselves, however exotic and unusual they may appear, but they cannot be part of the same classification system. This is in other words not a semantic problem. From a semantic viewpoint every category is reasonable, or at least not impossible, but something still makes the classification system as such impossible. This is because every individual part of a classification system must be juxtaposed to the other parts of the system. This means, that they must share a ground upon which to be juxtaposed. In this case there is no way that each of the individual categories can share the same ground. In some cases because they are candidates to the same place in that ground, and in some cases, because they leave no space for the other categories, and again in some cases because they eliminate the very ground. Foucault puts it this way;

”Borges adds no figure to the atlas of the impossible; nowhere does he strike the spark of poetic confrontation; he simply dispenses with the least obvious, but most compelling, of necessities; he does away with the site, the mute ground upon which it is possible for entities to be juxtaposed”. (ibid:xvii).

Classification is not just a matter of arranging objects in systems, but more radically of rendering the cognition and recognition of those objects possible. Understanding the structure of the site ”...the mute ground upon which it is possible for entities to be juxtaposed”, is to understand the conditions underlying the production of knowledge. In order to understand a certain kind of knowledge one has then not to focus on the semantic properties of it, but on the underlying grammar.

The question is then in relation to ethnography and anthropology; how is the site of ethnographic classification structured? I am thus not interested in the semantic properties of ethnographic literature, but rather the underlying conditions that make ethnography possible, the structure of the site upon which ethnography produces knowledge about others or, in a more general sense, about otherness.

My hypothesis is, that language plays a central part in the creation of such a site, and this has profound consequences for the representation of other peoples and the representation of otherness in general. I will analyse the part played by language from a structural viewpoint because it is my thesis, that language is used both metaphorically and metonymically in ethnographic literature.

I will argue, that a theory of language is used as a metaphor in ethnographic literature. The theory of language is used metaphorically as a theory of culture. This is an analytical device, but one with profound theoretical implications.

The metonymic use of language, on the other hand, facilitates the mapping of a territorial structure of the otherness that ethnography thrives on. Language as metonym stems on the one hand from the idea, that a language is an attribute of a culture, or even the cause or the effect of a culture, and on the other hand it implies borders, which adds the possibility of mapping cultures.

Language as metaphor

It is considered an empiricist occupation to describe a culture, or the social structure of a people in an ethnographic monograph. But the relevance of such an ethnography, would hardly differ from a journalistic account of the place, were it not for the further implications inherent in the theoretically informed analysis. As such it can not be said to be entirely empirical.

Bu the empirical account remains the objective of an ethnographic monograph. ‘The proper study of mankind is mankind’ said the enlightenment philosophes. Not so for ethnography - the proper study of ethnography is a people . There have been different ways of defining a people, but the current and most pervasive is no doubt the ethnic group. This epithet covers the notion of a society with a corresponding culture. Now, anthropology may compare ethnic groups, and it may account for the interrelationship between ethnic groups, but contextuality is arranged in these entities. The problem is then how to define the borders of that object. The question becomes how to define who and what is part of this object, and who and what is not. Although seldom explicit it is almost always implied, that an ethnic group corresponds to a linguistic group [2]. The particular language may thus be said to be a metaphor of an ethnic group. Language gives credit to the idea of the objects existence. It serves well as an emblem under which to gather the elements that constitute contextuality.

But it is more than just an emblem. It is not irrelevant that it is language that serves as a metaphor. Language is par excellence the media of communication, knowledge and understanding. The metaphor then implies that this is also the case with the ethnic group. Within the ethnic group understanding is readily there, communication is possible and knowledge is shared. In relations between different ethnic groups understanding is impaired or, in some of the more radical anthropological theories, impossible.

This is, however, not all. Language also implies coherence, and excludes nonsense, incoherence, contradictions, and so on [3]. A language implies that it is a bounded whole within which understanding is implied by the context of the entire language - understanding a German word or a sentence in German is implied, not by Chinese, but by the rest of what is believed to make up the German language. You may or may not be able to translate the German term into Chinese, but the context is the German language. It is seen as a symbolic system, that in itself has no dynamic or contradictions. In a word - coherence. Language is seen as a kind of store, where there is an entire set of comprehensible sounds and sequences of sounds, that are readily accessible to the initiated because of his/her being in the entire language. To the uninitiated it is, alas, only sounds.

Within this idea of the culture of a particular group designated as an ethnic group, there has been a lot of diverse developments of theories. The most prominent strand takes its cue from Pierre Bourdieus and develops a notion of relations of power within the ethnic group. This is often to show how the distribution of knowledge within a culture is structured by social relations of power. This however does not touch on the basic view - namely that things are only meaningful within the culture. The dynamics arise only from the social implementation of culture, not from the inherent dynamics of culture. It still implies, with language as the underlying metaphor, that there is in theory at least, a complete store of possibilities within a whole, which is then mobilised for different purposes in different positions in the socially structured relations of power.

In sum, what I am arguing here is, that a language is often taken for granted per se to define the ethnic group, which constitutes the pivotal point [4] of anthropological theory, and which is presented in ethnographic monographs. Furthermore this ethnic group is seen as constituted by a social structure which is then connected with a unique cultural system, as unique as any language.

It does seem on a semantic level, to solve the problem, inherited from the enlightenment, as Gellner pointed out (op.cit.), of the relation between liberalism and the grand reason. The grand reason has become local. Each culture is in itself a unique and complete fulfilment of the grand reason. Liberalism is also served its justice in the fact, that each culture is reserved its right to its own grand reason - different from other cultures’ grand reason. Bourdieu and his many proselytes within anthropology may point out, that within each culture or symbolic system, there are many individuals still left out in the dark, because of their structural marginal position in the social relations of power. But this point cannot be made, unless one accepts, that there is, at least within the single culture, the possibility of being absolutely enlightened. To be left in the shadows presupposes the existence of the light. The cultural relativist (or, if you like, one may term Bourdieu a symbolic relativist) utopian idea of a neat bounded cultural whole is only impaired by the vicious forces of social relations of power.

The solutions of the problem is, however, only apparently, since it reappears on the local level. The grand reason still serves as the guiding light and liberalism is no longer a question of liberty, but of confinement within the local version of the grand reason. Liberalism is thus transformed into tolerance, and becomes not a question of ethics, but a question of seclusion and division. I shall come back to the political consequences of this in the conclusion. The analytical consequence is that language serves as metaphor in yet a third way. Since one of the fundamental characteristic of language must be that it is translatable (cf. Davidson 1985), it is implied that languages are in a way metaphors for each other. They are of the same kind. In the same way with cultures. They can, through the hard and meticulous work of the anthropologist, be translated into an understandable version to be communicated to the uninitiated. But since the metaphor is rather the philological than the linguistic idea of language it is on a semantic level that the translation occurs.

At this point, I want to make it clear, that I am not saying that the reason for this is that language serves as a metaphor in ethnographic literature. I am only saying that language as a metaphor in ethnographic literature facilitates the ideas, I am outlining. It serves as part of the structural configuration of the site, that anthropology produces its knowledge in (cf. Foucault, op.cit.).

This analytical move is logical, but it is not grounded in the empirical facts, as is argued by many anthropologists. It is rather a synthesis made in the course of the development of anthropology. It is thus deeply rooted in colonialism, and it is by no means the only possible move, let alone a necessary move. It is a common trick, however, of reducing an abstract phenomena to a supposedly concrete phenomena. A culture is not a thing, readily accessible for the empiricist observer. Language is in the sense, that its existence is not questioned. There are languages out there, and they are different from each other. In its striving to establish an object of its own, anthropology uses this analytical synthesis to make the culture of a people appear as an empirical phenomena. My point is then, that the idea of language as an empirical object with more or less definite borders underlies the conception of what defines the object of anthropology. Not in the sense, that anthropology studies languages, but in the sense that anthropology studies objects, the limits of which are defined, in part, by the limits of particular languages.

This leads me to the point of language as a metonym. I have touched upon this point implicitly in my analysis of language as a metaphor, but I shall try making it more explicit in the following.

Language as metonym

In a sense it could be said, that I am arguing that anthropology presents language as a metonym, as the main attribute of a culture, or sometimes as the cause of the culture (in some constructivist theories). I have shown above, that it can do so only after having created its object through the use of language as a metaphor. But having done so, language indeed implies metonymic relations between the objects of anthropology. Languages border each other. Where one language stops, the other begins. Each in its own right covers a kind of territory. And indeed this territory is often meant literally in a physical, geographical sense. In any way, it implies that the culture, or the language is a space, rather than a place. It is not just somewhere to be, or even to dwell, it is somewhere to inhabit. The individual members are thus ascribed, or even adscripted to the particular space of their language, subsequently their culture. They may be in physical Diaspora, and they may cross borders, but they have a real home [5]. This is what might be called the mapping of the territorially structure of otherness. The way that ethnography handles otherness is by dividing it into different complexes of contextuality, defined by reference to language. This, however applies only to the real others. Somehow it does not apply to the members of the western culturesphere. They, the real others are divided by culture along the lines of linguistic differences, whereas we are united by culture across the linguistic boundaries. It is evident in the sense that evn though there are a growing number of studies ‘at home’ in what is supposedly the anthropologists own culture, it is often implied, that one is studying a subgroup within the wider western culture. Western culture is the context of a group within that culture. And, interestingly these groups are only rarely defined by their language. In this sense it is still quite evident that Ethnography as a dicipline grew out of (especially british) colonialism. There are still (epistemic, if not intentionally political) reminiscences of the ‘divide and rule’ policy of british colonialism. The politics have changed. It is no longer the civilized worlds task to educate the savages. Indeed, as Gellner pointed out (op.cit) anthropology sees it as its task to explain that the savages are indeed highly sophistcated and live in complex, coherent cultures, that are as elaborate and logical (in their own terms), as our own. But though the politics have changed, the logic of the division persists.

Anthropology, on the other hand grew out of the enlightenment philosophy, but (again, especially in Britain) has become more and more synonymous with ethnography - which is in fact the development I am problematising here.

The making of boundaries that defines the object, allows the ethnographer/anthropologist to distance himself from his object. Although in a very concrete sense he is a part of his object (since the much celebrated method of ethnography is the ill-defined participant-observation, meaning that the ethnographer literally lives in the midst of his object), he can say that he is at a distance from his object, since they are defined by borders that do not include him. Thus language as a metonym implies, first of all, that the object has boundaries to be crossed, and, secondly, that the ethnographer starts from a point outside that boundary. Thus the subjectivist version of post-modern critiques of very diverse kinds all share the fundamental problem (although they differ in their way of handling the problem) that they have to deal with the idea of intruding on their object. What has come out of this has been a problematisation of the nature of representation, and especially of the possibility of representation of other peoples.

There has been a lot of debate on this problem, some bordering on extreme versions of idealism, but most of it seems caught in the problem precisely because it is seen as a question of crossing a border, perceived as being very real. That border is the border between the subject and the object, and the anthropological and ethnographic version of the post-modern debate has mainly focused on that aspect. It is problematising the ‘collapse’ of the subject and the object, but, and this is the very crucial point, without questioning the configuration of the object. It only questions the nature of the representation of that object. In a sense the post-modern debate, in anthropology at least, is not really post-modern, it is very modern, perhaps too modern, since it does not question the premises of its own critique - premises which are indeed not that far from the premises of what post-modernism claim to be criticising.

To sum up; Language as a metaphor arises a number of problems. It defines the nature of contextuality, not depending on the problem to be studied, but on an arbitrary relapse into a supposedly empirical phenomenon, that again implies a coherence within the object. This is because the labels defines context rather than the logic of the object.

The main problem of language as metonym is the reification of borders between cultures. The idea of borders between culture gives rise to a proliferation of debates and problems, in particular the problem of representation, that may be either irrelevant or misguided by the fact that the object in question is defined by a borderline, that in itself is either irrelevant or at least misguiding.

Conclusion

I am not, as it were, arguing against contextuality, but I wish to problematise the nature of the context. My worry is, that with the development in ethnography and anthropology, and especially the relationship between the two, contextualism has become an absolutist term that is shaped by, among other things, language as metaphor and metonym.

The development in which language has served as metaphor an metonym has meant that anthropology slowly has become more and more synonymous with ethnography. Dan Sperber has formulated the problem this way:

”Most anthropologists devote themselves to the meticulous description of a single people. They limit their theoretical ambitions to the improvement of classification or to short- or middle-range generalizations about ”bridewealth” in Africa or ”big-manship” in Melanesia. If challenged to spell out what they know about the human species in general, they will have little to say [...] Many anthropologists have even claimed that there is no such thing as a human nature, not realizing that they were thereby denying the very subject matter for anthropology” (Sperber 1985:64).

This is where I return to the point Bruce Kapferer makes, that through fieldwork anthropology can learn something about humanity. This may seen evident, but because of the preoccupation with contextuality has lead to absolutism, as outlined above, anthropology has been overtly concerned with explaining the coherence and internal logic of a radically different cultural system, rather than taking the lessons of the fieldwork as a particular instance of a universal human condition. This point relies on the cue Kapferer takes from Louis Dumont, that although we are all humans, we do not each incorporate the entire essence of humanity. We are each instances of the potentiality of being human. If one takes the full consequence of this point, as I indeed think one should, it would mean that anthropology is not about other peoples, but the science of mankind. Other peoples are not the object of anthropology, but rather the means through which anthropology can broaden, expand the study of mankind. The proper study of anthropology is mankind, ought in my view to be the revised slogan the discipline. There are not other peoples , there are people, and there is otherness, and the two need not coincide. Ethnography ought to inform anthropology, meaning that, to paraphrase Kapferer, that we learn from ethnography new things about what it also is to be human. But before that can be realised a revision of the site that facilitates the production of anthropological knowledge is demanded. In this paper, I hope to have provided part of that revision.

References:

Davidson, D.

1985: ”On the Very Idea of a Conceptual Scheme” in J. Rajchman & C. West (eds.): ”Post-analytic Philosophy” pp. 129-144. Columbia University Press, New York.

Foucault, M.

1992: ”The Order of Things”. Routledge.

Sperber, D.

1985: ”On Anthropological Knowledge”. Cambridge University Press.

[1] Paradoxically, it could be argued that a great many anthropologists continue to emphasise the inherent contradictions and incoherences in their own culture, while ignoring those of others.

[2] I am not implying that it is language exclusively that explicitly or implicitly defines the borders of an ethnic group. I am only saying, that it is so in most cases. In any case, the problem of absolutism persists no matter the metaphor, since the ethnic group defines the coherent whole that is the context, that can explain anything that goes on inside, and nothing that goes on outside.

[3] I am not a linguist, and I am not talking about the linguists’, or the philologists’ understanding of language. I am talking about the way it is comprehensed in anthropology.

[4] I deliberately use this term, since in some strands of anthropology the individaul is the nucleus, and again in others the starting point is the world system, etc. But the ethnic group remains the pivotal point.

[5] This incidentally is also quite helpful in studying nomadic peoples. They do not belong in a home as such, but the anthropologist can place them in that non-place that resembles home - the language and the culture.